## Cruxes I Have With Many LW Readers

There's a crux I seem to have with a lot of LWers that I've struggled to put my finger on for a long time but I think reduces to some combination of:

• faith in elegance vs. expectation of messiness;
• preference for axioms vs. examples;
• identification as primarily a scientist/truth-seeker vs. as an engineer/builder.

I tend to be more inclined towards the latter in each case, whereas I think a lot of LWers are inclined towards the former, with the potential exception of the author of realism about rationality, who seems to have opinions that overlap with many of my own. While I still feel uncomfortable with the above binaries, I've now gathered enough examples to at least list them as evidence for what I'm talking about.

## Example 1: Linear Algebra Textbooks

A few LWers have positively reviewed Linear Algebra Done Right (LADR), in particular complimenting it for revealing the inner workings of Linear Algebra. I too recently read most of this book and did a lot of the exercises. And... I liked it but seemingly less than the other reviewers. In particular, I enjoyed getting a lot of practice reading definition-theorem-proof style math and doing lots of proofs myself, but found myself wishing for more examples and discussion of how to compute things like eigenvalues in practice. While I know that's not what the book's about, the difference I'm pointing to is more that I found the omission of these things bothersome, whereas I suspect the other reviewers were happy with the focus on constructing the different objects mathematically (I'm also obviously making some assumptions here).

On the other hand, I've recently been reading sections of Shilov's Linear Algebra, which is more concrete but does more ugly stuff like present the determinant very early on, and I feel like I'm learning better from it.

I think one contributing factor towards this preference difference is that I tend to be more OK with unmotivated messiness if the messy thing is clearly useful for something but less OK slogging through a bunch of elegant but not-clear-what-it's-used-for build up. Another way to put this would be that I tend to like to get top-down view of a subject and then go depth-first afterwards, whereas others seem happy to learn bottom-up. I used to think this was because of my experience with programming where algorithms are pretty much always presented in
terms of their purpose and tend to be become messier as they get optimized for performance. I still like knowing the motivation for things, but I also accept that stuff that works for real applications often has a bunch of messiness. On the other hand, a lot of LWers are also programmers who are only now going deep on math and they seem to still be happy with the axiomatic math way of doing things. So having a programming background doesn't seem to correlate with my preferences that strongly...

What would be great would be if someone would chime in providing better hypotheses/explanations than the one I've given.

## Example 2: Scientists vs. Engineers as Role Models

Much of early LW content, the Sequences in particular, used scientists like Einstein and Feynman as role models in discussions (and also targets of criticism in fairness). While I love Feynman and Einstein too, I tend to also revere builders/engineers, such as John Carmack, Jeff Dean, and Konrad Zuse, but these types of people don't seem to get nearly as much praise on LW.

One explanation for this is that great but not necessarily thoughtful engineers can drive X-risk through their work. For example, here's a discussion where a few folks argue that AGI requires insight more than programming ability and explicitly mention needing Judea Pearl more than John Carmack. While this is a fair argument, I'm skeptical that it's the true rejection. Security mindset seems to be as common among engineers as it is among scientists given that most of the folks who participate in things like DefCon and work in computer security tend to be hardcore engineer types like Trammell Hudson. (In his original essay, Eliezer cites Bruce Schneier, definitely an engineer, as someone he trusts to have security mindset.)

Another potential explanation for this is that LW readers tend to like doing and learning about science (pure math included) more than doing engineering. It's plausible that people who were attracted to early LW/OB content and were compelled by arguments for X-risk tend to also prefer science to engineering.

## Conclusion

Unfortunately, I don't have some sort of nice insight to conclude this with. I don't think the differences between my and other LWers preferences are bad so much as an implicit thing that doesn't get discussed.

I am curious whether my dichotomies seem reasonably accurate to anyone reading this? And if so, do my hypotheses for them seem reasonable?

Showing 3 of 16 replies (Click to show all)

In an interesting turn of events, John Carmack announced today that he'll be pivoting to work on AGI.

9NaiveTortoise9moAt this point, I basically agree that we agree and that the most useful follow up action is for someone (read: me) to actually be the change they want to see and write some (object-level), and ideally good, content from a more engineering-y bent. As I mentioned in my reply to jimrandomh, a book review seems like a good place for me to start.
2Ruby9moCool. Looking forward to it!